The Biostatisties Core (Core C) will be responsible for providing support and methodological leadership to TTURC investigators in the design, development and implementation of all three major research projects (Phenotypes of Dependence, Steve Buka, PI; Endophenotypes: fMRI Study, Raymond Niaura, PI; Contextual Factors of Smoking; David Abrams, PI) as well as current (Candidate Genes, Susan Santangelo, PI) and future developmental projects. To this effect, Core C will provide statistical support and conduct methodological research related to studies of familiarity and assortative mating in nuclear families, stage-sequential models of substance use initiation and progression to dependence, and the construction and evaluation of risk profiles based on latent class analysis (LCA) and Item Response Theory (IRT) applied to multisymptom diagnostic data. In addition, Core C will (a) develop and maintain a primary study database, (b) implement an efficient, secure and accurate mechanism for on-line access to the database by Core personnel for purposes of study monitoring and analysis, (c) monitor and verify the quality of the data, including checks of accuracy, completeness and timeliness, implement procedures for guaranteeing the integrity and confidentiality of the database and (d)support investigators in both participating institutions (Brown, Harvard) in their efforts to access the primary database and to perform on-site data analyses.